 mlemanczyk
|
aa4ff799f3
|
Small optimizations
|
2025-10-16 19:46:25 +02:00 |
|
|
mlemanczyk
|
07395830a5
|
Unnecessary cleanup commented out
|
2025-10-16 18:13:28 +02:00 |
|
|
Marek Lemańczyk
|
e53bb6ba43
|
Track small-prime remainders with cycle steppers
|
2025-10-16 17:18:27 +02:00 |
|
|
Marek Lemańczyk
|
fb08a61c66
|
Restore pooled small factor buffers
|
2025-10-16 17:18:19 +02:00 |
|
|
Marek Lemańczyk
|
949b5f891c
|
Return windowed Montgomery odd powers to pooling
|
2025-10-16 04:52:22 +02:00 |
|
|
Marek Lemańczyk
|
3e7ee6ab46
|
Restore pooled buffers in partial factorization
|
2025-10-16 04:52:17 +02:00 |
|
|
Marek Lemańczyk
|
9c474fc241
|
Use stack buffers for small composite collections
|
2025-10-16 00:59:00 +02:00 |
|
|
Marek Lemańczyk
|
3e7bffc812
|
Refactor heuristic pooling to cut GC pressure
|
2025-10-16 00:37:01 +02:00 |
|
|
Marek Lemańczyk
|
0a23502441
|
Clear heuristic candidate list without try-finally
|
2025-10-15 23:57:46 +02:00 |
|
|
Marek Lemańczyk
|
51730bcba3
|
Reduce GC pressure in divisor checking pipeline
|
2025-10-15 23:43:27 +02:00 |
|
|
Marek Lemańczyk
|
571b5a9938
|
Replace divisor scan UInt128 usage with thread-static GpuUInt128
|
2025-10-15 22:48:36 +02:00 |
|
|
Marek Lemańczyk
|
9ab87a8b9b
|
Streamline factor cache lease usage in CPU tester
|
2025-10-15 21:55:30 +02:00 |
|
|
Marek Lemańczyk
|
64c3721db4
|
Streamline divisor factor pooling for CPU hot path
|
2025-10-15 21:50:11 +02:00 |
|
|
Marek Lemańczyk
|
c99aebdf68
|
Restore divisor cycle fallback and enforce session disposal guard
|
2025-10-15 21:29:24 +02:00 |
|
|
Marek Lemańczyk
|
763714307f
|
Eliminate GC allocations from CPU divisor scan hot path
|
2025-10-15 20:48:42 +02:00 |
|
|
Marek Lemańczyk
|
b0ce7152f7
|
Pool factor cache entries to keep CPU factoring allocation-free
|
2025-10-15 20:48:33 +02:00 |
|
|
Marek Lemańczyk
|
34fcd62c23
|
Eliminate GC pressure from divisor cycle factoring
|
2025-10-15 19:32:53 +02:00 |
|
|
Marek Lemańczyk
|
c8f8132db7
|
Align divisor pooling with review feedback
|
2025-10-15 19:32:45 +02:00 |
|
|
Marek Lemańczyk
|
e44662413a
|
Use thread-static pools in CPU divisor testing
|
2025-10-15 18:10:39 +02:00 |
|
|
Marek Lemańczyk
|
0b831b5a23
|
Reduce PartialFactor GC pressure
|
2025-10-15 04:39:46 +02:00 |
|
|
mlemanczyk
|
2f85636a2a
|
Tiny cleanup
|
2025-10-15 03:21:38 +02:00 |
|
|
mlemanczyk
|
d3695a3dc3
|
Pools made thread static. Various small optimizations. "in" parameter added to read-only method parameters
|
2025-10-15 02:17:13 +02:00 |
|
|
mlemanczyk
|
7c71d5165e
|
TODO added
|
2025-10-13 21:16:03 +02:00 |
|
|
mlemanczyk
|
e9373dd8c9
|
PollardRho extended to allow running with 10_240 threads. Configuration made static to prevent unnecessary GC pressure.
|
2025-10-13 20:44:15 +02:00 |
|
|
mlemanczyk
|
313286f70d
|
Configuration adjusted
|
2025-10-13 20:13:44 +02:00 |
|
|
mlemanczyk
|
5a4a5e4d98
|
Heuristic defaults restored.
|
2025-10-10 22:48:46 +02:00 |
|
|
mlemanczyk
|
ee2df92a15
|
Performance improvement to decrease startup time.
|
2025-10-10 22:48:28 +02:00 |
|
|
Marek Lemańczyk
|
c9ae56a5ec
|
Remove default prime order overloads
|
2025-10-10 21:51:41 +02:00 |
|
|
mlemanczyk
|
64c08f9b49
|
We'll resolve the check when divisor cycle equals p. Tiny cleanup
|
2025-10-09 00:18:48 +02:00 |
|
|
Marek Lemańczyk
|
adac9ff5f9
|
Compact by-divisor candidate filtering
|
2025-10-08 23:34:35 +02:00 |
|
|
Marek Lemańczyk
|
9a81f5f685
|
Enable parallel by-divisor scanning threads
|
2025-10-08 22:45:53 +02:00 |
|
|
Marek Lemańczyk
|
85a98eb009
|
Reduce GPU remainder transfers for divisor filtering
|
2025-10-08 22:28:36 +02:00 |
|
|
Marek Lemańczyk
|
c45b3951c3
|
Align CPU by-divisor scan with cached divisor filters
|
2025-10-08 20:15:26 +02:00 |
|
|
mlemanczyk
|
6dc26336c6
|
Remainder check adjusted to PrimesGenerator.
|
2025-10-08 19:56:53 +02:00 |
|
|
Marek Lemańczyk
|
85634da6eb
|
Filter CPU divisor checks by admissible residues
|
2025-10-08 19:23:09 +02:00 |
|
|
Marek Lemańczyk
|
6918bd28ff
|
Make by-divisor GPU kernels reuse read-only operands
|
2025-10-08 18:47:16 +02:00 |
|
|
Marek Lemańczyk
|
cd1debe9d6
|
Various micro- and medium- optimizations
|
2025-10-07 22:14:34 +02:00 |
|
|
Marek Lemańczyk
|
48127845b5
|
GPU errors won't go silently, so that we can analyze and fix them.
|
2025-10-07 21:57:27 +02:00 |
|
|
Marek Lemańczyk
|
31e95e5a8b
|
Merge pull request #120 from mlemanczyk/codex/refactor-cycle-calculation-methods-for-mersenne
Streamline by-divisor cycle equality checks
|
2025-10-05 21:33:10 +02:00 |
|
|
Marek Lemańczyk
|
16181903a1
|
Streamline by-divisor cycle equality checks
|
2025-10-05 21:32:40 +02:00 |
|
|
Marek Lemańczyk
|
bd8686b86b
|
Unused parameter removed
|
2025-10-05 19:14:42 +02:00 |
|
|
Marek Lemańczyk
|
581da24b2f
|
Restore GPU incremental scan logic behind exception
|
2025-10-05 17:39:31 +02:00 |
|
|
Marek Lemańczyk
|
20120f5bc0
|
Fix odd power table updates in GPU heuristics
|
2025-10-05 16:55:27 +02:00 |
|
|
Marek Lemańczyk
|
ce1355c749
|
Compilation errors fixed
|
2025-10-05 13:13:47 +02:00 |
|
|
Marek Lemańczyk
|
6b1ebf01a6
|
Replace inline array in Montgomery table for GPU support
|
2025-10-05 06:42:38 +02:00 |
|
|
Marek Lemańczyk
|
abab307da4
|
Fix GPU-safe Montgomery window exponentiation
|
2025-10-05 06:35:18 +02:00 |
|
|
Marek Lemańczyk
|
26cc574020
|
Treat Pollard Rho success as the GPU primality check
|
2025-10-05 06:26:09 +02:00 |
|
|
Marek Lemańczyk
|
b6b372bc59
|
TODO moved to the code to concentrate on mirroring heuristic cycle algorithm on GPU
|
2025-10-05 02:17:08 +02:00 |
|
|
Marek Lemańczyk
|
0bc97937aa
|
Restore GPU-backed UInt128 pow2 Montgomery helper
|
2025-10-05 02:13:31 +02:00 |
|
|
Marek Lemańczyk
|
72dd69bd53
|
Align GPU pow2 heuristics with Montgomery helper
|
2025-10-04 22:40:27 +02:00 |
|